Energies 2014, 7, 1376-1392; doi:10.

3390/en7031376
OPEN ACCESS

energies
ISSN 1996-1073
www.mdpi.com/journal/energies
Article

Design and Control of a Multi-Functional Energy Recovery Power

Accumulator Battery Pack Testing System for Electric Vehicles
Bo Long 1, Ji Hyoung Ryu 2, Shin Teak Lim 2 and Kil To Chong 2,*
1

School of Mechanical, Electronic, and Industrial Engineering, University of Electronic Science and
Technology of China, Chengdu 611731, Sichuan, China; E-Mail: longbo1978@gmail.com
Departments of Electronics & Information Engineering, Chonbuk National University, Jeonju 567,
Korea; E-Mails: lovesure@hanmail.net (J.H.R.); mirr46@jbnu.ac.kr (S.T.L.)

* Author to whom correspondence should be addressed; E-Mail: kitchong@chonbuk.ac.kr;

Tel: +82-63-270-2478; Fax: +82-63-270-2394.
Received: 3 January 2014; in revised form: 10 February 2014 / Accepted: 21 February 2014 /
Published: 5 March 2014

Abstract: In this paper, aiming at the energy loss and harmonic problems in the
conventional power accumulator battery pack testing system (PABPTS), an improved
multi-functional energy recovery PABPTS (ERPABPTS) for electric vehicles (EVs)
was proposed. The improved system has the functions of harmonic detection, suppression,
reactive compensation and energy recovery. The ERPABPTS, which contains a bi-directional
buck-boost direct current (DC)-DC converter and a bi-directional alternating current (AC)-DC
converter with an inductor-capacitor-inductor (LCL) type filter interfacing to the AC-grid,
is proposed. System configuration and operation principle of the combined system are
discussed first, then, the reactive compensation and harmonic suppression controller under
balanced grid-voltage condition are presented. Design of a fourth order band-pass
Butterworth filter for current harmonic detection is put forward, and the reactive
compensator design procedure considering the non-linear load is also illustrated. The proposed
scheme is implemented in a 175-kW prototype in the laboratory. Simulation and
experimental results show that the combined configuration can effectively realize energy
recovery for high accuracy current test requirement, meanwhile, can effectively achieve
reactive compensation and current harmonic suppression.
Keywords: energy recovery; power accumulator battery pack testing system (PABPTS);
harmonic detection; Butterworth filter

Energies 2014, 7

1377

1. Introduction
The power accumulator battery pack testing system (PABPTS), which is used for evaluating the
performance of high power density power accumulator battery packs (such as lead-acid battery,
ultra-capacitor, lithium-ion battery, Fe battery, etc.), has been widely adopted in many battery
companies. The PAPBTS can offer instantaneous charging and discharging current test experiments,
which are especially useful in battery powered electric vehicles (EVs). When the EV is in climbing or
in downhill running mode, braking kinetics could be controlled transferring to and from the power
battery pack (PBP), thus, the braking energy can be effectively recycled. In this circumstance, the
battery pack is frequently charged and discharged, and for this reason, the PABPTS is a high accuracy
testing system which can be used to evaluate the characteristics and performance of the produced PBP.
The basic requirements for PABPTS have been indicated in [1,2], regarding the key issues for PABPTS,
much of the prior work has been done in the laboratory. A systematic design and control strategy for
an energy recovery PABPTS using a two-level converter have been discussed in [2], compared with
the small power output PAPBPTS in which the power transistors work in linear mode, the power
transistors in the proposed power circuit work in switching mode, which moreover, has the merits of
energy recovery and high power-output capability [2]. The power circuit of the ERPABTS has been
explained and implemented. In [3], modeling and control of an energy recovery PABPTS (ERPABPTS)
under charging mode are presented. Based on that, in order to further reduce the volume and
current harmonic of the fundamental frequency transformer in current PABPTS, design and control of a
transformer-less high power output battery testing system using a three-level neutral point clamped (NPC)
inverter is presented in [4], parameter design with inductor-capacitor-inductor (LCL) filter interface
and power flow control strategy for ERPABPTS have been put forward in [5].
Based on aforementioned work [25], considering that harmonics and reactive power exist due to
the non-linear loads (such as saturated transformers, arc furnaces and semiconductor switches), the
presence of harmonics and reactive power in the grid are harmful, and this will bring about additional
power losses and cause some malfunctions of other instruments sharing the same grid. To prevent the
inflow of harmonic and reactive currents and improve the operating ability of the transmission system,
a kind of Flexible AC Transmission System (FACTS) has been proposed. Moreover, in recent years,
the static Var compensator (SVC) and active power filter (APF) also have been put forward to
overcome the harmonic and reactive problems in grid-connected systems. Therefore, how to unify and
realize energy recovery, harmonic detection and reactive compensation in one combined ERPABPTS
are the main objectives of this paper.
The state of the art in harmonic detection and reactive compensation can be categorized into the
frequency and time-domain ones. Abundant literatures on this topic can be found in [6,7]. A brief
summary and some additional references about the most common methods related with frequency-domain
detection can be found in [8,9] for discrete Fourier transform (DFT) and in [10] for sliding DFT.
A second-order single-phase harmonic compensation method for a wind power system using a
proportional and resonant (PR) controller as a feed-forward compensator have been elaborated in [11].
Harmonic detection methods in the time-domain under stationary reference frame filtering usually
adopt fundamental notch filtering, second-order generalized integrators (SOGIs) and sinusoidal signal
integrators (SSIs) [12]. Instantaneous power theory (p-q power theory) and neural-network-based

Energies 2014, 7

1378

techniques [13,14] are also used for harmonic detection. Recently, a three-phase selective harmonic
detection method based on a cascaded delayed signal cancellation (CDSC) phase-lock-loop (PLL) is
discussed, it uses -frame CDSC operation to extract the harmonic of interest, and the proposed
scheme could completely eliminate the undesired harmonics [15].
Based on all the work mentioned above, we know that much of this work has been done in the
fields of reactive compensation, harmonic detection and suppression, but few papers have been found
illustrating a way to unify harmonic suppression, reactive compensation and energy recovery in
one system. This paper will combine harmonic detection with instantaneous reactive compensation in
an ERPABPTS. Battery testing system requires fast responses with minimum overshot, and combing
direct power control (DPC) with harmonic suppression and reactive compensation would be more
beneficial for high resolution battery testing system. Therefore, we first put forward and analyze the
operation principle of the proposed multi-functional ERPABTS, then, considering the reactive power
and harmonic distortion problems, the corresponding control strategy will be given and implemented.
This paper is organized as follows: the operation principles of ERPABPTS considering non-linear
loads are reviewed in Section 2; a compound control strategy considering harmonic detection,
suppression, reactive compensation and energy recovery is analyzed in Section 3; simulation verification
of the unified control scheme is presented in Section 4; experimental results of the proposed scheme are
given in Section 5; and some conclusions and future trends are presented in Section 6.
2. Operation Principle and Control Structure of ERPABPTS Including Non-Linear Load
2.1. Power Circuit of ERPABPTS
The detailed power circuit topology for the ERPBPTS is shown in Figure 1. The power circuit is
composed by a non-isolating cascaded connected power converter. A direct current (DC)-DC boost
chopper is responsible for wide range of constant and high accuracy charging and discharging
current test, a three-level NPC voltage source converter (VSC) is used for energy recovery. An LCL-type
current filter is used as an interface between the power converter and the alternating current (AC)-grid.
Large value and volume of DC-link capacitors have to be used to maintain DC-link voltage constant.
Discharging energy is temporarily stored in the DC-link capacitors. Therefore, the system is decoupled,
which means that the two cascaded power converters could be controlled independently. Compared with
conventional ERPABPTS, which use a step-down fundamental-frequency transformer as the interface,
the proposed power circuit topology has much smaller volume and higher energy recovery efficiency.
A non-linear load which is commonly used in the factory at the grid-side is also shown in Figure 1.
It is needed to note that in many circumstances a non-linear load (such as a diode-based rectifier)
might exist at the point of common coupling (PCC), thus, the power grid will be influenced greatly,
by factors such as line heat loss, noise, etc. Some of the low power instruments may even malfunction.
Therefore, a multi-functional ERPABPTS which can perform reactive compensation, energy recovery
and high resolution battery pack test is greatly needed.

Energies 2014, 7

1379

Figure 1. Power circuit configuration for energy recovery power accumulator battery
pack testing system (ERPAPBTS) considering a non-linear load. DC: direct current;
and VSC: voltage source converter.

Cd 1

ea
eb
ec

Cd 2

idis

iinv

ig

2.2. Operation Principle of ERPABPTS

The operation principle of the ERPABPTS can be described as when performing discharging
current test experiments, the DC-DC converter functions as a boost chopper, T6 switches ON and OFF
with pulse width modulation (PWM). High accuracy discharging current is obtained via the bypass
diode of T5, which gives rise of the DC-link capacitor voltage. Without being properly controlled, the
power transistors can be damaged by high-voltage break down. Integrating the energy of DC-link
capacitors to the AC-grid is a way to solve this problem. In this paper, the temporary energy is released
to the power grid through a VSC. Energy balanced control between DC-DC and DC-AC converter
is the most important issue for implementation. In this way, the discharging energy can be recovered
effectively, which, at the same, can realize high resolution discharging current tests. Current filter,
over-current protection devices, and fast-acting fuses are also used for protection. Whenever a non-linear
load exists on the grid-side, the reactive power component is automatically detected and compensated
by the power converter, which can be regarded as a unified grid-connected inverter and an APF.
In the following sections, we will first elaborate the operation principle of a DC-DC converter for
ERPABPTS, then, a compensation control strategy considering the non-linear load will be analyzed.
3. Charge and Discharge Control of the DC-DC Chopper for EPABPTS
In the combined system, the DC-DC converter regulates the discharging current of the PBP, as a
testing system, wide-range, high-resolution discharging and charging currents with fast response and
minimum steady state error are the basic requirements. This function is used for mimicking the
instantaneous startup and driving process in EVs. To improve the dynamic response, and decrease the
steady-state error of the DC-DC power converter, an optimal non-linear controller and the topology
of discharging part for battery pack were proposed in Figure 2, in which, iref is the reference
discharging current, id is the feedback discharging current, vbat and vDC-link are terminal voltage of
power accumulator battery pack and DC-link capacitors, respectively. L is the inductance for boosting.

Energies 2014, 7

1380

The small signal average model, stability of DC-DC boost chopper, and its corresponding non-linear
digital current controller in continuous current mode (CCM) have been illustrated in [1618] and
implemented in our previous work [2,19]. The experiments included high resolution internal resistance
tests, and constant discharging current tests which can be a double closed loop system (an outside
voltage loop and an inner side current loop), our work in this paper will concentrate on harmonic
detection and reactive compensation.
Figure 2. Control strategy of the DC-DC boost chopper in discharging mode.

vdc-link

idc-link
Cd 1
L

Cd 2

vbat

Req

idis

iref +

idis

vbat vdc-link

4. Harmonic Detection and Compensation Control for ERPABPTS Using Instantaneous

Power Theory
4.1. Harmonic Detection and Compensation Control of ERPABPTS
Figure 3 illustrates the proposed control diagram block scheme of the energy recovery grid-connected
PABPTS. In order to obtain the harmonic current component of the non-linear load, expressions for
the active and reactive power components when using instantaneous power theory are shown in
Equation (1):

p = 2 ( ed id + eq iq )

q = 3 ( e i e i )
d q
q d

(1)

In Equation (1), the voltage orientation control (VOC) method is used, which means ed = |E|, eq = 0,
Equation (1) can be rewritten as:

Energies 2014, 7

1381

p = 2 ( ed id )

q = 3 ( e i )
d q

(2)

The grid-side voltage transformation block is used to acquire the equivalent voltage under
synchronous rotation coordinates. Similarly, the load current transformation block is adopted to
obtain the equivalent current under synchronous rotation coordinates. However, Equation (2) itself
contains fundamental frequency and higher-order frequency active and reactive power components.
Thus, the higher-order frequency reactive power component should be distracted from total reactive
power component. A low-pass-filter (LPF) is needed in this circumstance. This procedure can be
explained as:
ph = p pf

qh = q qf

(3)

where pf and qf stand for the active and reactive power components, respectively; ph and qh denote the
total harmonic power components. After we get the higher-order reactive component, an inverse Clark
transformation block is used to acquire the active and reactive current components. Assuming that the
phase-currents of the non-linear load are asymmetrical, and contains higher-order current harmonics
which can be described as:

i
2 I n sin(nt + n )
=

a
n =1

=
i
b 2 I n sin(n t + n )
3

n =1

ic = 2 I n sin(n t +
+ n )
3

n =1

(4)

Based on Equation (4), after Clark transformation, the equivalent phase current under stationary
reference frame would be:

I n sin(nt + n )

i

n =1

i = 3

I n cos(nt + n )
n =1

(5)

Considering that the grid-side voltage might be unbalanced, and can be written as:

e
=
a 2 en sin(nt + n )
n =1

eb = 2 en sin(n t + n )
3

n =1

ec = 2 en sin(n t +
+ n )
3

n =1

(6)

Energies 2014, 7

1382

Similarly, the expression of phase voltage under stationary reference frame using the Clark
transformation would be:

en sin(nt + n )

e
n =1

e = 3

en cos(nt + n )
n =1

(7)

Combing Equation (5) with Equation (7), according to the instantaneous power theory, the active and
reactive power component of the system can be concluded as:
p = e i + e i

p = e i e i

(8)

Substituting Equations (5) and (7) into Equation (8), the instantaneous active and reactive power
component would be:

p
3
en I n sin 2 (nt + n )

n =1

p = 3 e I cos 2 (nt + )

n
n
n

n =1

(9)

Thus, active and reactive power component at fundamental frequency would be:

pf = 3 e1 I1 sin 2 (t + 1 )

2
qf = 3 e1 I1 cos (t + 1 )

(10)

Since the peak value of grid-voltage e = 3E1 , the fundamental frequency phase current derived by
using the inverse Clark transformation would be:

+
sin(
n

)
1
iaf

cos(t + 1 ) sin(t + 1 ) p
1
2

C23
ibf =
q = 2 I1 sin(t 3 + 1 )
2

+
+
sin(
t

)
cos(
t

)
3
E

1
1
1

cf
2
sin(t +
+ 1 )
3

(11)

According to Equation (11), the higher-order phase current (iah, ibh and ich) needs to be compensated,
and can be deduced by:
iah = ia iaf

ibh = ib ibf
i = i i
ch c cf

(12)

where iah, ibh and ich are the reactive and harmonic current needed to be compensated. Assuming that
ipa, ipb and ipc are the inverter-side phase current and are needed to be injected to the power grid,
the reference current for the VSC at inverter side should be:

Energies 2014, 7

1383
iaref = iah + ipa

ibref = ibh + ipb

icref = ich + ipc

(13)

where iaref, ibref and icref are inverter-side reference current and are determined by the outer closed- loop
of DC-link voltage. This procedure is shown in Figure 3.
Figure 3. Block diagram of the proposed reactive compensation and harmonic
suppression scheme. LPF: low-pass-filter.
ea

ea
eb
ec

iLa

p f

id

i*af

iah
+ ibh

*
ibf

iLb

iLc

pf

iq

i*cf
+

vbat
iL
D

id

iq
ed

ich +

iaref

ibref

icref

i pa
i pb
i pc

eq

In Figure 3, ea, eb and ec are the phase voltage of the AC-grid, respectively. e and e are the
grid-voltage under stationary reference frame. iLa, iLb and iLc are the phase current of the non-linear load.
The three-phase currents are firstly transformed into two-phase stationary reference frame, and then
are changed into two-phase synchronous rotating reference frame to get DC-current component
(iLd and iLq). According to the instantaneous power theory, active and reactive power component
(p and q) under stationary reference frame can be acquired. After being filtered by the designed
band-pass filter, the DC-current active and reactive power component (pDC and qDC) under synchronous
reference frame are derived. The reactive and harmonic power component can be obtained by
subtracting p and q from pDC and qDC.
4.3. Band-Pass Butterworth Power Filter Design

In order to acquire the higher order active and reactive power component, two second-order
band-pass butt-worth filters are designed in this paper. Since the fundamental frequency of the grid is
50 Hz, hence, the center-frequency is chosen to be 50 Hz, the passing-band frequency is set to be 20 Hz,
by using sptools block in Matlab, the continuous time transfer function of the filter can be designed as:

Energies 2014, 7

1384

G (s) =

1.579 104 s 2
s 4 + 251.3s 3 + 2.132 105 s 2 + 2.481102 s + 9.741109

(14)

Sampling time is chosen as Ts = 5 s, then, the discrete-time transfer function of the filter in
Equation (14) would be:

G(z) =

1.973107 z3 1.974 107 z2 1.971107 z +1.972 107

z4 3.999z3 + 5.996 z2 3.996z + 0.9987

(15)

Based on Equation (15), the Bode diagram of the designed band-pass filter would be derived.
From Figure 4, we can see that the magnitude is 0.193 dB at the fundamental frequency 50 Hz
(321 rad/s), the delay margin is 0.01 sec at (314 rad/s), and the magnitude has a big attenuation near
the center frequency point, phase delay at the fundamental frequency is close to zero. As a result,
the designed filter could satisfy our requirements.
Figure 4. Bode diagram of the proposed band-pass filter.
Bode Diagram
0
System: G1
Frequency (rad/s): 312
Magnitude (dB): -0.193

Magnitude (dB)

-20

-40

-60

-80
180

Phase (deg)

90

0
System: G1
Phase Margin (deg): 180
Delay Margin (sec): 0.01
At frequency (rad/s): 314
Closed loop stable? Yes

-90

-180

10

10

10

10

Frequency (rad/s)

4. Simulation Verification

To describe and verify the validity of the proposed scheme, simulation verifications are carried out.
For this purpose this work is implemented by co-simulation of Matlab/Simulink and Powersim 9.0.
Powersim has the advantages of fast computation and convergence capability, and is considered as a
professional power electronics simulation software. A simcoupler block is set up for the dynamic data
exchange between the two environments. The schematic diagram of the proposed control technique in a
distribution grid are shown in Figure 5, which contains a full-bridge diode-based rectifier with a resistance
and a capacitance in parallel, and an inductance load at the DC-link. Parameters of the diode-based full
bridge rectifier can be referenced in Table 1.

Energies 2014, 7

1385

Figure 5. Block diagram of the proposed compensation scheme. PI: proportional-integral.

20m

Nonlinear Load
A

0.0003

Source

0.47m

0.5

60m

Ia

iaf

PI

Ib

PI

Ic

PI

Hz

Table 1. Parameter specifications adopted for simulation. AC: alternating current; and
LCL: inductor-capacitor-inductor.
Component
Non-linear load

AC-grid

LCL filter

Part name
Inductance
Capacitance
Resitance
Lineto line voltage
Frequency
Internal inductance
Internal resistance
Grid side inductance
Capacitance
Reactive damping resistance in series
Inverter side inductance

Values
0.3 mH
470 F
0.5
110 V
60 Hz
0.1 mH
0.5
0.5 mH
2 F
0.1
2 mH

The following steps are performed to validate the performance of the unified ERPABPTS:

Energies 2014, 7

1386

(1) Without compensation, the non-linear load is only powered by the AC-grid. If without active
current component injection, the compensation current keeps track of the harmonic current.
The phase current of the non-linear load and the AC-grid are shown in Figure 6a,b. It can be
clearly seen that the load current is distorted and the total harmonic of the phase current on
the load is 13.46%. From Figure 6b, we know that the fifth order and seventh order harmonic
current exist in the loads. The harmonic current are 10 A at fifth order and 5 A at seventh
order, respectively.
Figure 6. Phase current and spectrum of the non-linear load: (a) load current; and (b) spectrum
of the load current. THD: total harmonic distortion.
100

150

I(i_sa)
i_La

load current

THD=13.46%

100

80

I(i_sa) (A)

I(i_sa)

50

60

40

-50

20

-100

-150
0.00

0.02

0.04

0.06

0.08

0.10

Time (s)

(a)

200

400

Frequency (Hz)

600

800

(b)

(2) To compare the performance of the proposed scheme and conventional scheme, simulation
results for the non-linear load with and without reactive compensation and harmonic detection
are illustrated, firstly. Then, by using the proposed fourth-order low-pass Butterworth filter
with sampling time Ts = 20 s, cut-off frequency fc = 55 Hz, the proposed reactive
compensation and harmonic suppression scheme are provided. The parameters of the
proportional-integral (PI) current controller in Figure 5 are set as: kp = 2, time constant: T = 0.0001.
Transfer function of the current controller is:
G (s)

(110
= 2

+ s)

(16)

Figure 7a gives the time response of the phase-current in the non-linear load before and after
integration of the ERPABPTS into the power grid without injecting the discharging current. An LCL-type
filter is adopted to acquire high quality grid-connected current. Compared with the L- and LC-type filters,
it has much smaller volume and current ripple, yet it brings the problems of resonance; hence, a passive
damping resistance of 0.1 is used in this paper. Compensation point is started at t = 50 ms,
from Figure 7b, the grid side phase-current after compensation becomes more sinusoidal, and its total
harmonic distortion (THD) reduced from 13.46% to 1.89%, compared with Figure 7b, the current
amplitudes of the fifth-order and seventh-order are counteracted. Moreover, the current amplitude at
fundamental frequency changes from 85 A to 80 A.

Energies 2014, 7

1387

Figure 7. Time response and spectrum of the grid-side phase current before and after
integration of the non-linear load: (a) load current; and (b) spectrum of the load current.
200
150

I(i_sa)
i_La

Start compensation

100

THD=1.89%
80

100

60

isa (A)

I(i_sa) (A)

50
0
-50

40

-100

20

Grid-side phase current

load current

-150
-200
0.00

0.02

0.04

0.06

0.08

0.10

Time (s)

0
0

200

400

Frequency (Hz)

(a)

600

800

(b)

(3) In terms of energy recovery, the energy discharged from the battery pack should be recovered
to the power grid. Thus, based on the simulation work done in Step 2, active current component
should be injected for modulation, which is demonstrated in Figure 8, where iaref, ibref and icref
are the additional currents needed to be recovered to the power grid.
Figure 8. Multi-functional ERPABPTS using the proposed scheme.

Ib

V
fcmpa
Ia

Ic

PI
iaref

PI

icmpb
Ib

V35

ibref

icmpc

PI
icref

Ic

Hz

Figure 9a illustrates the time response of the grid-side phase current without compensation,
with compensation and with reactive compensation. It can be seen that the THD of the grid current is
13.46% before compensation and 1.89% after compensation. From Figure 9a, we can see that when an
active current component iq is given by 100 A at 150 ms, the grid-side current amplitude reduced
from 117 A to 88 A, which means that the additional current 29 A is used to compensate the reactive
and harmonic current. Figure 9b shows the time response of the compensation current (icmp), which is
derived by the difference between the load current (iLa) and fundamental frequency current (ifa), which

Energies 2014, 7

1388

is calculated by using the proposed fourth-order filter. From Figure 9b, we can see that when the
three-phase voltage source inverter starts compensation at t = 50 ms, the phase current of power source
becomes highly sinusoidal, and has a THD of 1.89%. In Figure 9c, when t = 150 ms, the additional
active current component iaref, ibref and icref are added before PI controller in Figure 9. The VSC starts to
inject active current to the AC-grid, the active current (iaref, ibref and icref) reference is determined by
DC-link voltage to maintain constant. Hence, a unity power factor control scheme with reactive
compensation is established. Figure 9d gives the time response of the phase-current in the non-linear
load and the filtered current using the designed fourth-order Butterworth filter. It can be obviously
seen that after being filtered by the designed band-pass filter, the fundamental frequency current of the
non-linear load can be successfully extracted from the load current iLa.
Figure 9. Time response of the (a) grid-side current; (b) compensation current;
(c) inverter-side current; and (d) load current.
200

400
300
200

icmp

I(i_sa)

150

active current injection point

compensation start

100
50

icmp (A)

I(i_sa) (A)

100
0
-100

-50

-200

-100

with non-linear load

-300
-400
0.00

compensation start point

-150

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

-200
0.00

0.20

Time (s)

0.02

0.04

0.06

0.08

0.10

0.12

(a)

(b)

400

300

active current injection

iLa
ifa

300

200

200

start compensation
100

100

iLa (A)

I(i_inva) (A)

0.14

Time (s)

-100
-100

-200
-200

-300
0.00

-300

0.05

0.10

0.15

Time (s)

( c)

0.20

0.25

-400
0.00

0.05

0.10

0.15

0.20

0.25

Time (s)

(d)

5. Experimental Section

A multi-functional inverter, which is used for a grid-connected energy recovery Li-ion type power
battery testing system experimental platform, is set up to verify the proposed design. A microcontroller
(DSP-TMS320LF2808, Texas Instrument, Dallas, TX, USA) is used for fast computation and

Energies 2014, 7

1389

implementation. The hardware configuration setup of the power circuit is shown in Figure 10a. The Li-ion
battery used for experiment is presented in Figure 10b. Parameters of the power transistor,
inverter-side filter, DC-DC converter and AC-DC converter are listed in Tables A1 and A2 in the
Appendix. The composite control scheme can be referenced in the simulation section. Figure 10c gives
the phase-current of the power-grid before and after connecting the non-linear load with and without
reactive compensation. Similar with the simulation results, after reactive compensation, the phase-current
of the power-grid becomes more sinusoidal. Thus, a unified multi-functional ERPABPTS is established,
which is more applicable for a battery test. Figure 10d gives the waveform of the DC-link current,
which is generated due to the DC-DC converter in discharging mode, a step reference current of 400 A
is given at 50 ms, from Figure 10d, we may know that after a slight current overshot with the peak
value of 420 A, the instantaneous peak current is caused due to the existence of DC-link capacitance.
The feedback current keeps track of the reference current very well, hence, requirements on high
resolution, high power output and wide range of voltage and current test experiment is realized,
which demonstrated the correctness and utility of the proposed scheme.
Figure 10. Hardware setup and experimental results of the proposed scheme: (a) power circuit
configuration of the system; (b) Li-ion battery pack; (c) phase-current of the power source;
(d) active current component injecting to the AC-grid; and (e) time response of the
discharging current from the battery pack.

(a)

(b)
500

200

I(i_sa)

iqrefin

150
400

100
300

iqrefin

I(i_sa)

50
0

200

-50
-100

100

-150
-200
0.00

0.02

0.04

0.06

0.08

Time (s)

( c)

0.10

0.12

0.14

0.16

0
0.00

0.01

0.02

0.03

0.04

0.05

Time (s)

(d)

0.06

0.07

0.08

0.09

0.10

Energies 2014, 7

1390
Figure 10. Cont.
500

400

Discharging current

ibat (A)

300

200

100

0
0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

Time (s)

( e)
6. Conclusions

A unified harmonic suppression, reactive compensation and energy recovery power accumulator
battery pack test system is proposed. The topology and control strategy proposed in this paper can
eliminate harmonics generated by the nearby non-linear loads and realize reactive compensation.
The proposed reactive power and harmonic detection scheme can effectively detect the fundamental
frequency active and reactive power components. Compared with a conventional battery pack test system,
the unified control scheme improves and extends the performance of the combined system. A
TMS320LF2808-based DSP controller using the proposed method has been developed, and is
implemented in a 175-kW prototype in the laboratory. The results show that the THD is reduced from
13.46% to 1.89% in grid-connected condition, which demonstrates the good performance for harmonic
elimination and the reactive compensation scheme proposed in this paper.
Since many ERPABPTS sharing the same AC-grid are used in battery companies, it is suggested
that the future work might focus on ways of solving coordination control and energy management
problems existing in multi-ERPABPTS systems.
Acknowledgments

This work was simultaneously supported by the Fundamental Research Funds for the Central
Universities of China (No. ZYGX2012J095), China Postdoctoral Science Foundation Funded Project
(2013M542266), and also supported by the National Research Foundation of Korea (NRF) grant
funded by the Korea government (MEST) (No. 2013-009458) and (No. 2013-068127). The authors
would like to thank all the reviewers for their advices and suggestions on improving this paper.
Conflicts of Interest

The authors declare no conflict of interest.

Energies 2014, 7

1391

References

1.

2.
3.
4.
5.
6.

7.
8.

9.
10.
11.
12.

13.

14.

15.
16.

Zhang, L.; Kang, W.; Jia, L.; Zhao, M.; Xu, R. Control of bidirectional current source
SVPWM converter in the power accumulator battery testing system. In Proceedings of the
Asia-Pacific Power and Energy Engineering Conference (APPEEC), Chengdu, Sichuan, China,
2831 March 2010; pp. 14.
Long, B.; Xian, Z.; Lim, S.T.; Chong, K.T. Design of energy recovery power battery pack
testing system. Int. Rev. Electr. Eng. 2012, 7, 46724683.
Long, B.; Chong, K.T. Modeling and direct power control of energy recovery power battery
testing system under charging modeA new approach. Int. Rev. Electr. Eng. 2012, 7, 59936004.
Long, B.; Lim, D.; Chong, K.T. Design of transformer-less high power output battery testing
system using three-level NPC inverter. Int. Rev. Electr. Eng. 2012, 7, 61046115.
Long, B.; Chong, K.T. Parameter design and power flow control of energy recovery power
accumulator battery pack testing system. J. Electr. Eng. Technol. 2013, 8, 742753.
Zaveri, N.; Chudasama, A. Control strategies for harmonic mitigation and power factor correction
using shunt active filter under various source voltage conditions. Int. J. Electr. Power Energy Syst.
2012, 42, 661671.
Asiminciaei, L.; Blaabjerg, F.; Hansen, S. Detection is keyHarmonic detection methods for
active power filter applications. IEEE Ind. Appl. Mag. 2007, 13, 2233.
McGrath, B.P.; Holmes, D.G.; Galloway, J.J.H. Power converter line synchronization using a
discrete Fourier transform (DFT) based on a variable sample rate. IEEE Trans. Power Electron.
2005, 20, 877884.
Gonzalez, S.A.; Garcia-Retegui, R.; Benedetti, M. Harmonic computation technique suitable for
active power filters. IEEE Trans. Ind. Electron. 2007, 54, 27912796.
Jacobsen, E.; Lyons, R. The sliding DFT. IEEE Signal Process. Mag. 2003, 20, 7480.
Jeong, H.-G.; Lee, J.-H.; Lee, K.-B. A 2nd order harmonic compensation method for wind power
system using a PR controller. J. Electr. Eng. Technol. 2013, 8, 507515.
Yuan, X.M.; Merk, W.; Stemmler, H.; Allmeling, J. Stationary-frame generalized integrators for
current control of active power filters with zero steady-state error for current harmonics of
concern under unbalanced and distorted operating conditions. IEEE Trans. Ind. Appl. 2002, 38,
523532.
Qian, L.; Cartes, D.; Li, H. Experimental verification and comparison of MAFC method and
D-Q method for selective harmonic detection. In Proceedings of the 32nd Annual Conference on
IEEE Industrial Electronics, IECON, Paris, France, 610 November 2006; pp. 2530.
Cirrincione, M.; Pucci, M.; Vitale, G.; Miraoui, A. Current harmonic compensation by a single-phase
shunt active power filter controlled by adaptive neural filtering. IEEE Trans. Ind. Electron.
2009, 56, 31283143.
Wang, Y.F.; Li, Y.W. Three-phase cascaded delayed signal cancellation PLL for fast selective
harmonic detection. IEEE Trans. Ind. Electron. 2013, 60, 14521463.
Jung, Y.S. Small-signal model-based design of digital current-mode control. IEE Proc. Electr.
Power Appl. 2005, 152, 871877.

Energies 2014, 7

1392

17. Al-Mothafar, M.R.D.; Hammad, K.A. Small-signal modelling of peak current-mode controlled
buck-derived circuits. IEE Proc. Electr. Power Appl. 1999, 146, 607619.
18. Veerachary, M. Modelling and analysis of cascade step-down converters. IEE Proc. Electr.
Power Appl. 2005, 152, 4150.
19. Chan, C.Y. A nonlinear control for DC-DC power converters. IEEE Trans. Power Electron.
2007, 22, 216222.
Appendix
Table A1. Power transistor specifications adopted for the three-phase neutral point
clamped (NPC)-VSC inverter. IGBT: insulated gate bipolar translator.
Components
IGBT
Fast recovery diode
Capacitor

Part name/manufacturer
SKM400GB128D/Semikron (Berlin, Germany)
SKKD75F12/Semikron
YDK Technologies Company (Tokyo, Japan)

Rating values
1,200 V400 A
1,200 V75 A
450 V5,000 F

Table A2. Specifications adopted for the proposed ERPABPTS under discharging mode.
PBP: power battery pack.
Elements
PBP
AC Power grid

LCL Filter

DC-DC Converter

DC-AC Converter

Parameters
Battery terminal voltage Vbat
Battery type
Grid voltage (line to line rms) Vg
Line frequency fn
Grid inductance Lgrid
Inverter side inductor Linv
Grid side inductor Lg
Filter capacitance C
Nominal power
Inductor for boost chopping L
Switching frequency fs
Dead time td
Nominal power Pe
Two series DC-link capacitor CDC
Initial DC-link capacitor voltage Vc0
DC-link voltage reference VDC_ref
IGBT switching frequency finv
Dead time td

Values
240 V
Li-ion
380 V
50 Hz
1 mH
1 mH
0.5 mH
4.7 F
175 kW
4 mH
5,000 Hz
2 s
175 kW
16,000 F
500 V
900 V
2,000 Hz
2 s

2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/3.0/).